Nice! Does that work alongside reading in via chunks and writing via row_groups? If I have a 500GB CSV will it work?